ComfyUI/comfy_extras/nodes_sag.py

171 lines
5.9 KiB
Python
Raw Normal View History

import torch
from torch import einsum
import torch.nn.functional as F
import math
from einops import rearrange, repeat
import os
2024-05-14 22:02:27 +00:00
from comfy.ldm.modules.attention import optimized_attention
import comfy.samplers
# from comfy/ldm/modules/attention.py
# but modified to return attention scores as well as output
2024-05-14 22:02:27 +00:00
def attention_basic_with_sim(q, k, v, heads, mask=None, attn_precision=None):
b, _, dim_head = q.shape
dim_head //= heads
scale = dim_head ** -0.5
h = heads
q, k, v = map(
lambda t: t.unsqueeze(3)
.reshape(b, -1, heads, dim_head)
.permute(0, 2, 1, 3)
.reshape(b * heads, -1, dim_head)
.contiguous(),
(q, k, v),
)
# force cast to fp32 to avoid overflowing
2024-05-14 22:02:27 +00:00
if attn_precision == torch.float32:
2023-12-15 06:28:16 +00:00
sim = einsum('b i d, b j d -> b i j', q.float(), k.float()) * scale
else:
sim = einsum('b i d, b j d -> b i j', q, k) * scale
del q, k
if mask is not None:
mask = rearrange(mask, 'b ... -> b (...)')
max_neg_value = -torch.finfo(sim.dtype).max
mask = repeat(mask, 'b j -> (b h) () j', h=h)
sim.masked_fill_(~mask, max_neg_value)
# attention, what we cannot get enough of
sim = sim.softmax(dim=-1)
out = einsum('b i j, b j d -> b i d', sim.to(v.dtype), v)
out = (
out.unsqueeze(0)
.reshape(b, heads, -1, dim_head)
.permute(0, 2, 1, 3)
.reshape(b, -1, heads * dim_head)
)
return (out, sim)
def create_blur_map(x0, attn, sigma=3.0, threshold=1.0):
# reshape and GAP the attention map
_, hw1, hw2 = attn.shape
b, _, lh, lw = x0.shape
attn = attn.reshape(b, -1, hw1, hw2)
# Global Average Pool
mask = attn.mean(1, keepdim=False).sum(1, keepdim=False) > threshold
ratio = 2**(math.ceil(math.sqrt(lh * lw / hw1)) - 1).bit_length()
mid_shape = [math.ceil(lh / ratio), math.ceil(lw / ratio)]
# Reshape
mask = (
mask.reshape(b, *mid_shape)
.unsqueeze(1)
.type(attn.dtype)
)
# Upsample
mask = F.interpolate(mask, (lh, lw))
blurred = gaussian_blur_2d(x0, kernel_size=9, sigma=sigma)
blurred = blurred * mask + x0 * (1 - mask)
return blurred
def gaussian_blur_2d(img, kernel_size, sigma):
ksize_half = (kernel_size - 1) * 0.5
x = torch.linspace(-ksize_half, ksize_half, steps=kernel_size)
pdf = torch.exp(-0.5 * (x / sigma).pow(2))
x_kernel = pdf / pdf.sum()
x_kernel = x_kernel.to(device=img.device, dtype=img.dtype)
kernel2d = torch.mm(x_kernel[:, None], x_kernel[None, :])
kernel2d = kernel2d.expand(img.shape[-3], 1, kernel2d.shape[0], kernel2d.shape[1])
padding = [kernel_size // 2, kernel_size // 2, kernel_size // 2, kernel_size // 2]
img = F.pad(img, padding, mode="reflect")
img = F.conv2d(img, kernel2d, groups=img.shape[-3])
return img
class SelfAttentionGuidance:
@classmethod
def INPUT_TYPES(s):
return {"required": { "model": ("MODEL",),
"scale": ("FLOAT", {"default": 0.5, "min": -2.0, "max": 5.0, "step": 0.1}),
"blur_sigma": ("FLOAT", {"default": 2.0, "min": 0.0, "max": 10.0, "step": 0.1}),
}}
RETURN_TYPES = ("MODEL",)
FUNCTION = "patch"
CATEGORY = "_for_testing"
def patch(self, model, scale, blur_sigma):
m = model.clone()
attn_scores = None
# TODO: make this work properly with chunked batches
# currently, we can only save the attn from one UNet call
def attn_and_record(q, k, v, extra_options):
nonlocal attn_scores
# if uncond, save the attention scores
heads = extra_options["n_heads"]
cond_or_uncond = extra_options["cond_or_uncond"]
b = q.shape[0] // len(cond_or_uncond)
if 1 in cond_or_uncond:
uncond_index = cond_or_uncond.index(1)
# do the entire attention operation, but save the attention scores to attn_scores
2024-05-14 22:02:27 +00:00
(out, sim) = attention_basic_with_sim(q, k, v, heads=heads, attn_precision=extra_options["attn_precision"])
# when using a higher batch size, I BELIEVE the result batch dimension is [uc1, ... ucn, c1, ... cn]
n_slices = heads * b
attn_scores = sim[n_slices * uncond_index:n_slices * (uncond_index+1)]
return out
else:
2024-05-14 22:02:27 +00:00
return optimized_attention(q, k, v, heads=heads, attn_precision=extra_options["attn_precision"])
def post_cfg_function(args):
nonlocal attn_scores
uncond_attn = attn_scores
sag_scale = scale
sag_sigma = blur_sigma
sag_threshold = 1.0
model = args["model"]
uncond_pred = args["uncond_denoised"]
uncond = args["uncond"]
cfg_result = args["denoised"]
sigma = args["sigma"]
model_options = args["model_options"]
x = args["input"]
2024-01-10 09:08:43 +00:00
if min(cfg_result.shape[2:]) <= 4: #skip when too small to add padding
return cfg_result
# create the adversarially blurred image
degraded = create_blur_map(uncond_pred, uncond_attn, sag_sigma, sag_threshold)
degraded_noised = degraded + x - uncond_pred
# call into the UNet
(sag,) = comfy.samplers.calc_cond_batch(model, [uncond], degraded_noised, sigma, model_options)
return cfg_result + (degraded - sag) * sag_scale
2023-12-18 22:03:32 +00:00
m.set_model_sampler_post_cfg_function(post_cfg_function, disable_cfg1_optimization=True)
# from diffusers:
# unet.mid_block.attentions[0].transformer_blocks[0].attn1.patch
m.set_model_attn1_replace(attn_and_record, "middle", 0, 0)
return (m, )
NODE_CLASS_MAPPINGS = {
"SelfAttentionGuidance": SelfAttentionGuidance,
}
NODE_DISPLAY_NAME_MAPPINGS = {
"SelfAttentionGuidance": "Self-Attention Guidance",
}